Fast rule identification and neighborhood selection for cellular automata.

School of Computer Science and Informatics, Cardiff University, Cardiff, UK.
IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics: a publication of the IEEE Systems, Man, and Cybernetics Society (Impact Factor: 3.78). 12/2010; 41(3):749-60. DOI: 10.1109/TSMCB.2010.2091271
Source: PubMed

ABSTRACT Cellular automata (CA) with given evolution rules have been widely investigated, but the inverse problem of extracting CA rules from observed data is less studied. Current CA rule extraction approaches are both time consuming and inefficient when selecting neighborhoods. We give a novel approach to identifying CA rules from observed data and selecting CA neighborhoods based on the identified CA model. Our identification algorithm uses a model linear in its parameters and gives a unified framework for representing the identification problem for both deterministic and probabilistic CA. Parameters are estimated based on a minimum variance criterion. An incremental procedure is applied during CA identification to select an initial coarse neighborhood. Redundant cells in the neighborhood are then removed based on parameter estimates, and the neighborhood size is determined using the Bayesian information criterion. Experimental results show the effectiveness of our algorithm and that it outperforms other leading CA identification algorithms.


Available from: Xianfang Sun, Jun 02, 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper a method is proposed which uses data mining techniques based on rough sets theory to select neighborhood and determine update rule for cellular automata (CA). According to the proposed approach, neighborhood is detected by reducts calculations and a rule-learning algorithm is applied to induce a set of decision rules that define the evolution of CA. Experiments were performed with use of synthetic as well as real-world data sets. The results show that the introduced method allows identification of both deterministic and probabilistic CA-based models of real-world phenomena.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Ahstract-A common image forgery method is copy-move forgery (CMF), where part of an image is copied and moved to a new location. Identification of CMF can be conducted by detection of duplicated regions in the image. This paper presents a new approach for CMF detection where cellular automata (CA) are used. The main idea is to divide an image into overlapping blocks and use CA to learn a set of rules. Those rules appropriately describe the intensity changes in every block and are used as features for detection of duplicated areas in the image. Use of CA for image processing implies use of pixels' intensities as cell states, leading to a combinatorial explosion in the number of possible rules and subsets of those rules. Therefore, we propose a reduced description based on a proper binary representation using local binary patterns (LBPs). For detection of plain CMF, where no transformation of the copied area is applied, sufficient detection is accomplished by ID CA. The main issue of the proposed method is its sensitivity to post-processing methods, such as the addition of noise or blurring. Coping with that is possible by pre-processing of the image using an averaging filter.
    IWSSIP 2014; 05/2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: When farmland is abandoned pasture is rapidly taken over by woody vegetation. As tree dispersal depends on the presence of a seed source nearby and other local conditions, and can be measured in discrete annual time steps, a Cellular Automata model (CA) is a natural fit for modelling this phenomenon. The model presented here is a stochastic CA, with a relaxed definition of neighbourhood. The aim is to explore sources of uncertainty in the model, and techniques for handling and visualising uncertainty. The results show that it is possible to realistically model vegetation change using CA, acknowledging and incorporating uncertainty.
    Proceedings of the SIRCNZ conference, Dunedin, New Zealand; 08/2013